Open pshipton opened 4 months ago
Possibly related to https://github.com/eclipse-openj9/openj9/issues/19678
Another one in an internal build [linux_390] SE80_GIT cmdLineTester_decompilationTests - lnxec414
Looks like 2/50 are failing in a 50x grinder
Seems something didn't release exclusive access.
@hzongaro @dmitripivkine not sure who should be looking at this.
I just realized the vmfarm machines where failures occur are all rhel6. There is one machine which shouldn't be enabled, and the other is a build machines where we will remove the test label soon. Maybe we shouldn't bother looking at those failures.
It is not clear why "hang" is detected. There is system core available in results. It was taken after GDB output was provided. From this core more than dozen threads have VM Access and running random places (publicFlags=20):
!j9vmthread 0xcafc00 publicFlags=0 privateFlags=1008 inNative=0 // main
!j9vmthread 0xcc5300 publicFlags=0 privateFlags=101a inNative=0 // JIT-SamplerThread
!j9vmthread 0xcc7900 publicFlags=0 privateFlags=101a inNative=0 // IProfiler
!j9vmthread 0xde0200 publicFlags=80100 privateFlags=2 inNative=0 // Common-Cleaner
!j9vmthread 0xdf2900 publicFlags=0 privateFlags=1a inNative=0 // Concurrent Mark Helper
!j9vmthread 0xe3fe00 publicFlags=0 privateFlags=20081a inNative=0 // Finalizer thread
!j9vmthread 0xeb2400 publicFlags=20 privateFlags=0 inNative=0 // Thread-0
!j9vmthread 0xeb4a00 publicFlags=20 privateFlags=0 inNative=0 // Thread-1
!j9vmthread 0xeb7000 publicFlags=20 privateFlags=400 inNative=0 // Thread-2
!j9vmthread 0xeb9600 publicFlags=20 privateFlags=400 inNative=0 // Thread-3
!j9vmthread 0xebbc00 publicFlags=20 privateFlags=400 inNative=0 // Thread-4
!j9vmthread 0xebe200 publicFlags=20 privateFlags=0 inNative=0 // Thread-5
!j9vmthread 0xec0900 publicFlags=20 privateFlags=400 inNative=0 // Thread-6
!j9vmthread 0xec2f00 publicFlags=20 privateFlags=400 inNative=0 // Thread-7
!j9vmthread 0xec5500 publicFlags=20 privateFlags=400 inNative=0 // Thread-8
!j9vmthread 0xec7b00 publicFlags=20 privateFlags=400 inNative=0 // Thread-9
!j9vmthread 0xeca100 publicFlags=20 privateFlags=0 inNative=0 // Thread-10
!j9vmthread 0xecc700 publicFlags=1000 privateFlags=0 inNative=0 // Thread-11
!j9vmthread 0xeced00 publicFlags=20 privateFlags=400 inNative=0 // Thread-12
!j9vmthread 0xed1300 publicFlags=1000 privateFlags=0 inNative=0 // Thread-13
!j9vmthread 0xed3a00 publicFlags=20 privateFlags=400 inNative=0 // Thread-14
!j9vmthread 0xed6000 publicFlags=1000 privateFlags=0 inNative=0 // Thread-15
This snapshot does not look as "hang".
I guess the machine is really slow and timeout has been triggered mid-run
It happens, sometimes these machines seem to get slow.
\https://openj9-jenkins.osuosl.org/job/Test_openjdk23_j9_extended.functional_s390x_linux_Nightly_testList_1/1 - ub20-390-6 cmdLineTester_decompilationTests_nongold_3
-Xcompressedrefs -Xjit -Xgcpolicy:gencon
Full output output.txt